HIV is adept in evading humoral immunity, a feature that likely contributes to its ability to replicate in the face of ongoing host immune responses. Although neutralizing antibodies are produced to viral envelope glycoproteins (Env), they are characteristically directed to hypervariable loops on gpl20 (V1/V2 and V3). These antibodies are in general type specific and easily circumvented by mutations. These variable loops also serve to protect domains on the gpl20 core, which include the highly conserved binding sites for CD4 and chemokine receptors required for cell entry. In order for broadly neutralizing antibodies to be produced, it is likely that these and other conserved domains will need to be targeted. A priority for HIV vaccine research is to develop Env-based immunogens that can elicit these antibodies. Our laboratory has developed a novel strategy for producing replication-competent variants of HIV-2 that lack the V1/V2 and V3 hpervariable loops. Although other groups have produced functional HIV-1 and SIV Envs without V1/V2, viruses lacking V3 have not been described. Given the importance of V3 in binding to and determining specificity for chemokine receptors, and of VI/V2 and V3 in protecting critical domains on the gpl20 core from humoral immune responses, we hypothesize that these minimized Envs will have novel structure/function properties, and that, while presenting exposed core domains in this physiological context, will elicit qualitatively different immune responses focused to biologically relevant and/or cryptic epitopes on the Env trimer. We propose 5 Specific Aims to further explore this HIV-2 model and to extend these findings to HIV-1 and SIV: 1) In our HIV-2 model, selection strategies will be used to derive and characterize viruses lacking V1/V2 and V3 in combination, with a goal of generating replication competent Env cores; 2) These HIV-2 Envs will be characterized in functional, biochemical, morphologic, and immunologic assays that dissect underlying mechanisms for their novel biological activities; 3) Compensatory mutations in gpl20 and gp41 that permit Env cores lacking hypervariable loops to be functional will be determined; 4) Information Aims 2 and 3 will be extended to similar studies of HIV-1 and SIV; and 5) The immunogenicity of virions derived in Aims 1 and 4 will be assessed in a small animal model and sera evaluated to determine if functionally critical but exposed epitopes can be rendered immunogenic by this approach. These novel HIV and SIV Env variants will )rovide new opportunities for studies of Env structure, function, immunogenicity and pathogenesis.